about
From GPU to TPU with Coding Agents: What Works, What Fails, and What It Costs? (arxiv.org)
1 point by matt_d 15 days ago | hide | past | pdf | discuss on HN

In plain words: An AI coding agent was asked to move ten machine-learning projects from NVIDIA chips to Google's, with each result checked for accuracy, hardware use, and speed. Eight succeeded, but 76% of runs failed—often silently, sometimes changing training and lowering accuracy.

Abstract

A coding agent can migrate a machine learning workload, report success, and still change how the model trains. Rising compute costs, limited accelerator capacity, and vendor dependence are pushing organizations toward alternatives to graphics processing units (GPUs); a key question is whether agent-generated migrations to tensor processing units (TPUs) can be trusted and at what cost. We build a dataset of 148 PyTorch ML research projects and use Claude Code to migrate ten of them from NVIDIA GPUs to Google Cloud TPUs. We independently assess whether each migration runs, preserves numerical behavior, executes on the intended hardware, maintains the project's goal, and delivers acceptable performance. We also examine failures, human effort, and costs. Eight projects meet our project-goal criterion, and migrations require 2.4 human interventions on average. Of 70 execution attempts, 53 fail (76%), and 26 of these failures (49%) raise no error. The most frequent cause is repeated compilation and other execution inefficiencies, while interactions with third-party libraries can silently change the computation: in one four-chip migration, the agent reports success despite a change in gradient clipping that reduces accuracy from 86.53% to 84.89%. Execution costs vary widely: in one workload, TorchTPU, a PyTorch execution backend for TPUs, trains 1.86 times faster and 25% cheaper per run than the GPU. For practitioners, these findings motivate independent reference runs, verification under the intended hardware configuration, and workload-specific acceptance and stopping criteria that account for both computational and human costs.

Hui Sun, Anderson Uchôa, Wesley K. G. Assunção
arXiv:2610.34422 · cs.SE, cs.DC, cs.LG · submitted Oct 25, 2026 · updated Oct 29, 2026
abstract · pdf · html · 12 pages, 2 figures, 2 tables

add comment on HN